Determining airplane boarding groups

The airplane boarding process is typically part of the critical path of an airplane’s turn-around. Since time at the gate is costly, it is essential for airlines to optimize this process. To address this challenge, numerous boarding policies have been developed and extensively studied in the existing literature. Typically, these policies divide passengers into boarding groups, with members of each group boarding the airplane at random. This paper builds on existing knowledge through the use of well established estimates of random aircraft boarding times. We rely on these analytical and empirical estimates to optimize the allocation of passengers to a predetermined number of boarding groups. Our approach incorporates a mathematical model and a simulation study, accounting for finite walking speeds. Theoretically, this comprehensive methodology provides the opportunity to determine novel boarding group configurations for a predetermined number of boarding groups. However, both estimates known from literature reveal that the well-known “window-middle-aisle” strategy performs best among all possible group assignments if passengers do not walk too slowly. If passengers walk slowly, “back-to-front” turns out to be best. The practical significance of our findings extends to tangible benefits for airlines as they provide new insights to optimize the boarding process both economically and operationally.